1. Field of the Invention
The present invention relates to an enhanced coupling arrangement for an optoelectronic transducer, and in particular, to an enhanced coupling arrangement that reduces and/or controls a gap between an optical coupler for an optical device, such as an optical fiber, and a surface emitting device, such as a vertical cavity surface emitting laser, for example.
2. Background Information
Computer and communication systems are now being developed in which optical devices, such as optical fibers, are used as a conduit (also known as a wave guide) for modulated light waves to transmit information. These systems include at least a light emitter, and an optical coupler that connects the optical device to the light emitter. A generic term of either a light emitter or a light detector is an xe2x80x9coptoelectronic transducer.xe2x80x9d
As an example, optoelectronic transducers convert electrical signals to or from optical signals; the optical signals carry data to a receiver from a transmitter at very high speeds. Typically, the optical signals are converted into, or converted from, associated electrical signals using known circuitry. Such optoelectronic transducers are often used in devices, such as computers, in which data must be transmitted at high rates of speed.
In order to transmit the optical signals, the light emitter is typically either a light emitting diode (LED) or laser emitter. Conventionally, a photodiode is used to receive the optical signals. Optical fibers, which collectively form a fiber-optic cable, may be coupled to the respective LED or laser, and the photodiode, so that the optical signals can be transmitted to and from other optoelectronic transducers, for example.
The optoelectronic transducers are normally located on either input/output cards or port cards that are connected to an input/output card. Moreover, in a computer system, for example, the input/output card (with the optoelectronic transducer attached thereto) is typically connected to a circuit board, for example a mother board. The assembly may then be positioned within a chassis, which is a frame fixed within a computer housing. The chassis serves to hold the assembly within the computer housing.
Typically, there are two different types of light emitters which may be utilized with optoelectronic transducers. These include, in general, edge emitters and surface emitters. Edge emitters typically have a light emitting portion which is located on an edge of a chip, and typically have an active area that may be, for example, half a micron by four microns in size, for a total area of about 2 square microns. In contrast, surface emitters, such as vertical cavity surface emitting lasers (VCSEL), conventionally have an active area that is substantially larger than the active area of an edge emitter. The active area of a surface emitter is typically around 20 microns in diameter, so as to provide for about, for example, 400 square microns of active area.
Moreover, and in contrast to a typical edge emitter, the conventional surface emitter has an active area that is surrounded by inactive portions. This allows further devices to be placed immediately adjacent to the surface emitter, using the inactive portions as bearing surfaces. Moreover, and in contrast to a typical edge emitter, surface emitters commonly include coatings, such as silicon dioxide or other nitrides, which are utilized for passivation purposes.
Further, with the conventional VCSEL, the light is emitted in a conical beam vertically from the surface of the chip. Furthermore, the conventional VCSEL allows for integrated two-dimensional array configurations. For example, the VCSELs can be arranged in a linear array, for instance 12 surface emitters spaced about 250 microns apart, or in area arrays, for example, 16xc3x9716 arrays or 8xc3x978 arrays. Of course, other arrangements of the arrays are also possible. Nevertheless, linear arrays are typically considered to be preferable for use with optoelectronic transducers, since it is generally considered easier to align the optical fibers which collect the light emitted from the VCSELs in a linear array, than in an area array. Moreover, it is also conventional not to arrange the VCSELs in any sort of array whatsoever, but instead utilize the VCSELs singly.
It is important to ensure that as much of the light emitted from the light emitters reaches the respective optical fibers. However, the light emitted from the light emitters always diverges. This divergence may cause some of the emitted light not to reach the optical fibers, thus decreasing the efficiency of the transmission. Thus, the gap that must be bridged between the emitter and the optical fibers must be carefully controlled.
Moreover, as the emitted light diverges, it becomes increasingly more difficult to align the emitted light with the respective optical fibers. That is, if the emitted light beam has a diameter that is smaller than a diameter of the respective optical fiber, there is a certain acceptable margin of error in the alignment process. For instance, the respective light emitter may be shifted slightly off-center relative to the respective optical fiber, with the emitted light still impinging completely upon the optical fiber. On the other hand, if the emitted light beam has a diameter that is, due to its divergence, the same as, or larger than the diameter of the respective optical fiber, any shifting of the light emitter away from center relative to the respective optical fiber will cause some of the emitted light to miss the optical fiber.
In order to reduce any misalignment between the optical fibers and the light emitters, so as to ensure that the emitted light does not partially or completely xe2x80x9cmissxe2x80x9d its intended target, the light emitters may be either actively or passively aligned. For a device to be actively aligned, the light emitter is typically turned on and the other elements aligned with the light emitter while the device is activated. By using this approach, each device produced is individually aligned. Obviously, this is not preferable if the devices are to be mass produced. However, when the positional tolerances are very small, active alignment may be the only acceptable way to ensure that the light emitters are aligned with the optical fibers, especially when there is a large divergence in the emitted light beams.
Alternatively, passive alignment techniques utilize jigs or other manual operations to align the light emitters to the respective optical fibers. Passive alignment techniques are less accurate that active alignment techniques, and thus work best when the positional tolerances are larger, that is, when some shifting of the light emitters relative to the respective optical fibers can be tolerated.
From the foregoing, it is clear that it is desirable to reduce the divergence of the light beam emitted from the light emitters as much as possible. One way to reduce the divergence of the light beam is to move the light emitter as close as possible to the optical fibers. However, due to the fragile nature of the light emitters, it is desirable that the surface of the light emitter does not directly contact the optical fibers, especially during the alignment process. Moreover, it is further desirable that the light emitters be fixed relative to the optical fibers, so as to maintain their relative positions to each other.
Referring to FIGS. 1 and 2, a known arrangement is illustrated, in which the light emitter 10, such as a VCSEL, is attached to a carrier 12, and the ends of optical fibers (not shown) are embedded within an optical coupler 14. In this conventional arrangement, the carrier 12, rather than the light emitter 10, is directly attached to the optical coupler 14. Moreover, it is understood that the light emitter is conventionally formed on a top surface of a chip. For explanatory purposes, both the light emitter and the associated chip are collectively referred to as light emitter 10.
The carrier 12 is typically molded or otherwise constructed from a thermally-conductive material, such as copper, to help dissipate heat generated by the light emitter 10. Moreover, the carrier 12 is conventionally provided with two contact surfaces 16, separated from each other by a well 18. The well 18 accommodates the chip having the light emitter thereon. Typically, the chip is adhered to the carrier 12 using an epoxy or other adhesive disposed in a bottom of the well 18. Moreover, an end face of the optical coupler 14 has a middle active region, in which the optical fibers are disposed. When the carrier 12 is attached to the optical coupler 14, the two contact surfaces 16 of the carrier will be disposed on either side of the middle active region of the optical coupler 14, so that the light emitter(s) 10 can be aligned with the optical fibers.
Moreover, since the carrier 12 is typically formed of copper, the carrier can be utilized to serve as a return electrical path for the light emitter 10. That is, an electrically conductive epoxy can be utilized to adhere the chip to the carrier 12, so that current passes from the light emitter 10 via the electrically conductive epoxy, into the carrier and subsequently transmitted elsewhere by utilizing, for example, electrical conductors on a flexible cable.
In order to assist in the accurate positioning and alignment of the light emitter 10, it is conventional to polish the various contacting surfaces (the end face of the optical coupler 14 and the contact surfaces 16 of the carrier) to be very flat, i.e., about two microns from peak to valley. Moreover, the chip has a known and defined thickness, and likewise has upper and lower surfaces that are very flat. Thus, when the chip is attached within the well 18, and the carrier 12 is attached to the end face of the optical coupler 14, a gap 20 having a known size will be formed between an upper surface of the light emitters 10 on the chip and the end face of the optical coupler. This gap 20 has conventionally been deemed necessary to prevent damage to the light emitter 10 during the alignment process.
However, each of the adjoining surfaces has a tolerance associated therewith. That is, the dimensions of the depth of the well 18, the thickness of the chip, the thickness of the epoxy or other adhesive used to attach the chip to the bottom of the well, and the surface flatness of the end face of the optical coupler 14 and the contact surfaces 16 of the carrier due to manufacturing tolerances, can vary. These tolerances are cumulative, so that the total accumulated tolerance is about 50 microns, with the resulting gap being between about 75 and 125 microns. This gap is relatively large, and will allow for a substantial divergence of the light beam, thus requiring complex active alignment techniques, and possibly reducing the amount of light that can be received by the optical fibers. Thus, there is a need for a coupling arrangement for an optoelectronic transducer, in which the gap is minimized so as to reduce the divergence of the emitted light.
Moreover, it is also known to utilize a lens in the coupling arrangement, to help focus the emitted beam down. However, using a lens is complicated and costly. Thus, there is a need for a coupling arrangement for an optoelectronic transducer, which reduces the divergence of the emitted light without requiring a lens to do so.
Moreover, it is also known to attach a chip directly (i.e., with no gap) to a mount which includes optical fibers therein (i.e., an optical coupler). For example, in U.S. Pat. No. 4,730,198 issued to Brown et al., it is disclosed that after aligning the chip with the optical fibers, the chip is held in place by application of a transparent epoxy or solder located between the chip and the mount. In this scenario, the chip is held against the end face of the mount in a so-called hard stop arrangement. That is, the chip is pushed against the mount until the two components are in direct contact with one another, at at least two different points. However, this arrangement is disadvantageous in that during alignment, the chip will be moved relative to the end face of the mount. Since the chip is in direct contact with the mount, it is possible that the surface of the chip will be damaged due to its engagement with the mount during its relative sliding motion. Thus, although this known arrangement allows for the gap between the chip and the mount to be reduced, because the alignment technique is performed prior to the application of the epoxy, it is very possible that the chip will be damaged during the alignment process. Thus, there is a need for a method and arrangement that will reduce the gap between the chip and the optical coupler, and which will allow the chip to be aligned without damage thereto.
It is, therefore, a principle object of this invention to provide an enhanced coupling arrangement for an optoelectronic transducer.
It is another object of the invention to provide an enhanced coupling arrangement for an optoelectronic transducer that solves the above mentioned problems.
These and other objects of the present invention are accomplished by the enhanced coupling arrangement for an optoelectronic transducer disclosed herein.
According to one aspect of the invention, a carrier is provided in which the conventional bonding surfaces have been either reduced in size or removed. A surface emitter, such as a VCSEL, is attached to the carrier, in a conventional manner. However, instead of attaching an optical coupler to the contact surfaces of the carrier, this exemplary aspect of the invention utilizes an adhesive, such as a urethane acrylate or epoxy resin, to couple the optical coupler to the surface emitter. In the exemplary embodiment, urethane acrylate (628 series), manufactured by Dymax ((trademark)) Corporation, of Torrington, Conn. was utilized. This urethane acrylate has a nominal viscosity of 650 cP (centipoise), and is cured with either UV or visible light, heat or activator.
Because of the viscosity of the adhesive, if a predetermined amount of force is applied between the optical coupler and the surface emitter, the adhesive will advantageously serve as a viscous, lubricating medium, so as to act as a bearing layer, and will prevent the surface of the surface emitter from coming in direct contact with the optical coupler. Thereafter, the optical coupler and the surface emitter can be aligned using either active or passive alignment techniques, without concern as to whether or not the optical coupler will damage the surface of the surface emitter. After alignment, the adhesive is cured, to affix the surface emitter to the optical coupler.
In a further aspect of the invention, the adhesive will include a plurality of small balls, each of which has a diameter of the preferred gap, for example, 10 microns or so. The balls advantageously ensure that if too much force is applied between the optical coupler and the surface emitter, some of the adhesive will remain in place therebetween. Moreover, the balls ensure that a parallel relationship will result between the active region of the optical coupler and the surface emitter. Furthermore, during alignment, the balls will tend to roll, so that the balls will serve as a bearing surface, which will not damage the surface emitter.
In a further aspect of the invention, the balls are made of a translucent material, such as glass or plastic, and will have an index of refraction which is substantially similar to the index of refraction of the adhesive. This ensures that the balls will not introduce additional scattering and potential loss of light.
Moreover, in a further aspect of the present invention, the balls may be comprised of a glass base and coated with a plastic coating. The glass base would serve as a relatively stiff base, but since the glass would be coated with the plastic, the glass would not destroy or degrade the surface of the surface emitter if pressed too hard. Instead, the plastic would serve as a buffer layer, which would act as a cushion. On the other hand, the underlying glass base would be substantially stronger than a pure plastic ball, which may become distorted if pressed too hard.
In another aspect of the invention, the balls may be disposed within a non-adhesive carrier. After the balls are used to establish the gap, an adhesive could then be applied to fix the surface emitter to the optical coupler.